customer experience teams

Abstract: Multimodal sentiment analysis (MSA) aims to infer sentiment by integrating linguistic, acoustic, and visual information. However, many fusion methods integrate modality-specific representations and cross-modal interactions into joint representations, making it difficult to determine whether predictive utility arises from a modality itself or from its interactions with others. More fundamentally, a modality's standalone predictive utility may not faithfully reflect its interaction-conditioned predictive utility. To address this issue, we propose I$^2$-KAN, an interpretable multi-pathway framework that explicitly maintains intra-modal and pairwise inter-modal information in separate structural pathways. Each pathway is refined by a Task-Decoupled Linear-Gated Denoising Bottleneck (TDL-GDB) to learn compact and task-relevant representations, followed by structured Kolmogorov-Arnold Network fusion that retains explicit pathway associations and supports multi-level interpretation. Extensive experiments on CMU-MOSI, CMU-MOSEI, and CH-SIMS v2 demonstrate strong overall performance. Ablation and interpretability analyses consistently show across the three benchmarks that low standalone predictive utility does not necessarily imply low interaction-conditioned predictive utility: a modality that appears weak in isolation can remain informative through pairwise interactions. These findings highlight the interaction-dependent nature of multimodal utility and suggest that modality utility should be characterized from both standalone and interaction-conditioned perspectives.

Abstract: Personalizing a language model requires knowing what a person values, which stays largely consistent across tasks. Yet most methods either learn task-specific preferences or place the user's profile in the prompt, where models struggle to act on it. We introduce VISTA (Value-Informed Steering and Targeted Acquisition), which infers a latent profile from users' (full or partial) survey answers, and maps the profile through a hypernetwork to a user-specific steering intervention in a frozen LLM to select answer choices or generate free-text responses. Given the same evidence, VISTA matches or beats in-context profiles on synthetic preferences (Food2D), the World Values Survey (WVS), and PrimeX, without adding prompt tokens. To reduce the cognitive burden of completing full surveys, VISTA also chooses which questions to ask and when to stop: on WVS it stops after 6% of the questionnaire while retaining 94% of the full-history gain, where standard adaptive testing retains at most 42%. The dominant steering direction is an interpretable value axis on which VISTA places unseen users accurately. Our results point toward pluralistic alignment that adapts to individuals at deployable cost.

Abstract: Taxonomies are the symbolic representations through which AI systems organize evidence, aggregate patterns, and answer questions over large document collections. Over customer feedback, the category tree decides how every record is counted and routed, which problems get seen, and which team owns them. Agentic harnesses now make it easy to generate a plausible-looking hierarchy, and such trees are checked today with generic, individually scoped checks: each name fits its description, sits under the right parent, and stays distinct from its siblings. We ask a more operational question: when is a generated hierarchy actually useful as a production taxonomy? We build six taxonomies over two proprietary feedback corpora (1,940 and 5,000 records): for each corpus, a production reference and two repeated runs of the same harness under identical inputs. All six pass every generic naming and structure check, and a deeper product-coverage check even prefers the generated trees. Yet in every generated tree at least 97.7% of leaf names merely restate an ancestor's name (13.9% and 2.9% in the references), and in one, three of every four records fall under multiple top-level categories. We introduce two families of whole-tree metrics: structural discriminators test whether a tree's shape was learned from the data or imposed by its generator; team partitionability tests whether branches split feedback into groups teams can own. Trees the generic checks rate as equally correct differ by 27 percentage points in cross-branch leakage, and only one of the two beats a random split. Surface plausibility is an insufficient measure of taxonomy quality: evaluation must measure the whole tree as well as each node.

Abstract: AI agents are increasingly being programmed to automate knowledge work over large collections of unstructured data. Such automation requires repeatability: when the underlying evidence is unchanged, the agent's categories, priorities, and counts should not shift materially between runs, even if each individual answer appears plausible. We introduce a repeat-run evaluation framework that aligns semantically equivalent categories and focuses on two operating metrics: theme churn, the normalized change in the returned category set, and volume disagreement, the change in counts for categories that persist. We evaluate three recurring customer-feedback tasks across eight frontier models, corpus sizes from 100 to 5,000 records, multiple prompts, and three execution designs: raw generation, taxonomy-free hierarchical decomposition, and a taxonomy-grounded agent (TGA) using persistent themes, subthemes, and record-level predictions. With Claude Opus 4.8 and the 1,000-record corpus fixed, TGA reduces theme churn by 86--88% relative to both raw generation and hierarchical decomposition, while matched-theme volumes have zero disagreement. The taxonomy-grounded agent is more stable than every raw model in the screen, remains more stable at each corpus size, and keeps this advantage when theme matching is made stricter or looser. Although evaluated on customer feedback, the framework targets repeated synthesis of unstructured corpora more broadly, including financial reports, legal documents, incident records, and scientific literature. Overall, these results show that taxonomy grounding produces more consistent and repeatable outputs for recurring knowledge work.

Abstract: Preference-based alignment methods such as reinforcement learning from human feedback (RLHF) and Nash learning from human feedback (NLHF) aggregate pairwise preferences to learn an LLM policy, but a natural goal is maximizing social welfare (average cardinal utility), which comparisons alone need not identify. Gölz, Haghtalab, and Yang (GHY) measure the gap by distortion: the worst-case ratio between the welfare of the best fixed lottery (distribution over responses) and of the learned lottery. They show NLHF is optimal when every user receives the same lottery. Account-based LLMs, however, have information about their users and can serve different lotteries to different people. We give an efficient algorithm, GLHF, that learns a single group-conditioned policy from one comparison per user. Under individual Bradley--Terry comparisons, GLHF asymptotically matches GHY's optimal population distortion bound simultaneously on every group in a prespecified, possibly overlapping collection, with sample complexity growing logarithmically in the number of groups and inversely with the smallest group mass. A sharper guarantee for groups with similar preferences approaches distortion of one when members share a feasible favorite response. In experiments using human coffee ratings and synthetic LLM-generated ratings, GLHF lowers distortion in every evaluated group and substantially reduces worst-group distortion relative to NLHF and other group-agnostic baselines.

Abstract: Large language models (LLMs) can infer user preferences from interaction histories and reviews, yet the rationales they generate may not reflect the information actually used for recommendation. A preference claim may be weakly supported by its selected evidence, or may have little effect on the final ranking. We refer to these two failures as the grounding-influence gap. We introduce PROVE-REC, a general framework for verifiable preference reasoning in LLM-based recommendation. Pass A converts the complete pre-target history into a compact preference proof consisting of positive and avoidance claims linked to selected evidence entries. Pass B predicts the next item using only the proof and its selected evidence, preventing the recommender from bypassing the reasoning path. To verify evidence-to-proof grounding, we compare the effect of masking selected evidence with masking a comparable control entry. To verify proof-to-recommendation influence, we remove a preference claim and measure the resulting decrease in the target item's ranking margin. A ranking-preservation objective further retains useful information from the complete history. Comprehensive experiments on wide-ranging real-world datasets demonstrate that PROVE-REC consistently outperforms strong sequential, generative, and LLM-enhanced baselines, with improvements of up to 7.45%. Controlled ablations confirm the effectiveness of the two-pass architecture and verification objectives. Moreover, PROVE-REC produces claims that are more strongly grounded in historical evidence and more influential to recommendation while preserving ranking quality.

Abstract: Post-deployment feedback from users can be a cost-effective, scalable, and representative means to monitor and improve generative AI systems and features. When implemented effectively, giving such feedback can increase users' engagement with and trust in GenAI systems. Government regulations and industry guidelines call for post-deployment user engagement, but there is little guidance on designing mechanisms that are usable for consumers and provide actionable input for product teams. We conducted a multi-phase study as a collaboration between academic researchers and eBay. Our benchmark evaluation of current industry approaches identified common issues including lack of discoverability, unclear terminology, and inattention to user value. Based on these findings, we developed best-practice recommendations and designed and tested a prototype feedback-collection tool. The tool aimed to provide users with an efficient, flexible, and positive feedback-giving experience, and provide product teams with rich data on performance and potential problems in a usable format.

Abstract: While recent Multimodal Large Language Models (MLLMs) have shown promise in affective computing, their reasoning capabilities are largely confined to short video clips with limited interactions. However, real-world emotions are not merely isolated instantaneous reactions but dynamic and cumulative processes deeply shaped by past experiences and ongoing events. To bridge this gap, we introduce LongEmoBench, a benchmark dedicated to emotion understanding and reasoning in long videos. It assesses progressive capabilities scaling from continuous scene interactions to complex episodic developments. Furthermore, we propose LongEmo, a novel memory-augmented agentic framework designed to tackle the immense challenges of long-range affective reasoning. LongEmo processes continuous video streams to construct an Event Memory Graph, explicitly modeling long-range dependencies and capturing emotional dynamics across discrete events. Given a question, the agent retrieves a query-relevant event stream from the graph, iteratively integrating multimodal memories and relational dependencies to deduce the final answer. Extensive evaluations of 17 representative methods reveal that they struggle significantly with emotion understanding and reasoning in long videos. In contrast, LongEmo achieves state-of-the-art performance, demonstrating the efficacy of its event-centric memory architecture.

Abstract: How can we personalize a shared expert library from a user's pairwise choices? Prior work can realize different reward trade-offs by merging reward-specialized experts, given a vector of trade-off weights. In practice, users can more naturally choose between outputs than specify numerical weights. The challenge is therefore to turn these choices into the coefficients required for merging, while accounting for ambiguity when feedback is limited. Our key idea is to treat the unknown reward weights as latent variables: infer a posterior over them from pairwise choices and reward-score differences, and use its mean directly as the merge coefficients. We instantiate this idea as Bayesian Pairwise Merging (BPM), whose posterior also characterizes which reward trade-offs remain plausible given the feedback. We evaluate BPM on radiology summarization, image captioning, and story generation, spanning text-to-text and image-to-text generation. With 100 feedback per simulated persona, BPM achieves macro decided win rates of 91.7%, 77.1%, and 64.3% against uniform merge. For six pairs of simulated personas, each prefers the model fitted to its own feedback, a pattern also observed in a human proof-of-concept. In simulations under BPM's model and prior, its nominal 90% intervals for temperature-scaled reward weights achieve task-averaged marginal coverage of 88.9% and 89.2% with only 10 and 25 comparisons, respectively. BPM thus enables personalization from pairwise feedback without per-user policy training, while characterizing the coefficient ambiguity left by limited feedback.

Abstract: Aspect-based sentiment analysis (ABSA) has largely turned to text generation. We show that competitive dimensional ABSA does not need it. Using Jev, a frozen model that answers typed questions with rubric scores, label probabilities, and yes/no judgments, we decompose all three tasks of SemEval-2026 Task III Track A into such decisions and align them with the annotation scheme through 488 coefficients fitted on CPU, with no text generation and no backbone tuning. On valence-arousal regression over ten corpora in six languages, the system reaches 1.0645 RMSE, the lowest aggregate error of any participating system. On triplet and quadruplet extraction, it reaches 52.09 and 44.06 continuous F1, above fine-tuned Llama-3.3-70B and GPT-OSS-120B baselines. Analyses and ablations show where the accuracy comes from: supervised calibration roughly halves the raw regression error, exact valence-arousal would add only 4.5 F1 to extraction, and the learned combination of span-boundary evidence, not any single signal, carries the extraction systems.

Abstract: We present a framework for evaluating and improving a large-scale, multi-agent shopping assistant in production, and report lessons from its use. Offline evaluation of such a system faces three obstacles. (i) A logged conversation cannot be replayed against a modified system, because a different response changes every turn that follows. (ii) The unchanged system itself varies from run to run. Its LLM components are stochastic, and in product search the available products, their prices, and the customer's personalization signals change. (iii) Aggregate quality scores combine distinct behaviors, so they show that quality has changed but not which behavior caused the change. Our framework addresses each obstacle in turn. For a reported behavior, an Evaluation Harness generates targeted assertions and a fixed cohort of customer scenarios. It then reproduces the behavior in a local instance of the assistant through grounded user simulation. Instead of replaying the log, the simulator writes new customer turns conditioned on the recorded messages and context. Repeated runs of the unchanged system form a stored baseline. An Improvement Orchestrator turns the assertion results into hypotheses, implements each as an isolated modification, and compares it with the baseline using paired percentile bootstrap intervals over scenario-level differences. When an investigation ends, the harness may propose revisions to future evaluations, subject to human approval and without altering past decisions. We report production investigations with this framework. Assertion profiles showed which positions of a product carousel a failure affected, and repeated runs distinguished a real improvement from run-to-run fluctuation. Audits of the evaluation itself found a judge that lacked the evidence it needed and a model setting that was configured but not applied.

Abstract: Voice agents are increasingly deployed in workflows where failed interactions can affect transactions, access, and other consequential outcomes, creating a need for reproducible and interpretable evaluation. We introduce Inquesto Score (IS), a protocol for measuring voice-agent reliability as the percentage of calls in a fixed, versioned evaluation population that achieve the caller's goal without a functional failure or worse. Rather than combining heterogeneous metrics, IS defines explicit failure events and severity levels and evaluates the deployed voice pipeline. Timing failures, including talk-over and delayed responses, are measured directly from audio, while semantic and state-dependent failures are evaluated using scenario predicates, tool traces, and a pinned open-model judge. Diagnostic views of behavior, acoustic robustness, identity handling, and speaker groups accompany the score without being combined into it. Inquesto Score v0.1 evaluates 30 scenarios, three acoustic conditions, four speaker groups, and 306 calls per agent across 13 configurations of a reference voice-agent system. Our evaluation shows that reliable measurement requires evidence beyond transcripts, explicit treatment of deployment conditions, and validation of the evaluators used to determine outcomes. We release the protocol, reference implementation, and evaluation records.

Abstract: Digital financial platforms make enrollment effortless and cancellation laborious. This asymmetry is a dark pattern that manipulates users who have already decided to leave. Existing work identifies such patterns after deployment, and regulators sanction them after harm, yet neither provides designers with a criterion for building interfaces that avoid manipulation. We model the provider as an adversary whose instrument is effort and express fairness as a constraint requiring that leaving never cost more than joining. Defining interaction cost over navigation steps, mandatory inputs, and confirmation prompts, we prove that this constraint holds for every assignment of effort weights if and only if no component of the exit flow exceeds its counterpart at entry. Fairness is therefore verifiable by counting rather than by estimating cognitive effort, and exact equivalence is unnecessary because exit legitimately requires fewer inputs than entry. We instantiate the model, together with invariants for visual parity and linguistic neutrality, in a mobile credit card prototype with parallel sign-up and cancellation workflows.

Abstract: Recent research on Multimodal Sentiment Analysis (MSA) has focused on learning from language, visual, and acoustic modalities with incomplete data to infer human sentiment. Most studies typically compensate for missing information by reconstructing modality features or designing complicated fusion mechanisms. However, these methods still suffer from spurious generation and noisy guidance due to the lack of high-level semantic grounding in partially observed multimodal evidence. To address these issues, we propose SemMSA, a latent semantic-aided framework that constructs rich sentiment-relevant semantics with LLMs, fully integrating with all modalities via anchor-free spectral alignment. It mainly consists of Cross-modal Semantic Refinement (CSR) and Cross-modal Spectral Alignment (CSA). Specifically, CSR first adaptively extracts visual and acoustic representations by corresponding adapters to form a unified multimodal prefix with language in the frozen LLM embedding space. It then iteratively produces continuous discriminative semantic states through a token-efficient latent refinement process without decoding explicit text. Next, CSA simultaneously aligns the refined semantics with all modalities by enhancing the dominant spectral component of their kernel Gram matrix. This captures global nonlinear dependencies among all representations without relying on a predefined anchor modality. In addition, an instance-level spectral separation constraint preserves cross-sample discriminability and mitigates representation collapse. Extensive experiments on SIMS, MOSI, and MOSEI benchmarks demonstrate that SemMSA achieves state-of-the-art performance.

Abstract: Increasing numbers of people now routinely interact with large language models (LLMs) across many aspects of life, including in the workplace, educational settings, and for personal activities. The pace at which these tools have been adopted across society in recent years has led to substantial shifts in the ways in which people approach tasks and access information and advice. As a result, it is important for researchers and policymakers to gain up-to-date evidence on how the public engages with these technologies, including what people typically use them for, the extent to which users trust the information and advice that LLMs provide, and how the public perceives the potential benefits and risks associated with their use. We surveyed a nationally representative sample of 2,002 adults in the UK. Participants were asked about their use of LLMs, including both practical applications and more personal and social forms of engagement, their trust in the information provided by these systems across a range of topics and compared with other common sources, and their attitudes towards the potential societal benefits and risks associated with these tools. Results show that while practical tasks remain the most common type of LLM use, many people now engage with these tools for personal support. Almost one third of regular users (31%) say they use LLMs for personal and emotional support, such as talking through problems and asking for help with decisions, while one quarter report interacting with LLMs for meaningful conversation. We also find that trust in LLM-generated information is relatively high, but that public attitudes towards LLMs are characterised by both optimism and concern. While over three quarters of our sample report feeling enthusiastic about the potential benefits of LLMs (77%), a majority also express concern about their potential risks (70%).

Abstract: AI companions can purportedly adopt racial personas, raising questions about how they represent identity and how users interpret these portrayals. We combined an algorithmic audit of race-coded AI personas with interviews with 12 companion users who interacted with a probe. Our audit revealed systematic differences, such as Asian-coded male personas receiving higher submissiveness scores than White counterparts, and Black, Hispanic, and Indigenous male personas receiving higher aggression scores than their White counterparts in open-weight models. Interviews revealed that participants envisioned AI companions as offering cultural familiarity and outside perspectives, but differed in which portrayals they considered meaningful or stereotypical. Some rejected overt racial signaling while still expecting culturally distinctive responses. Triangulating these findings with theory, we highlight how social norms and cultural expectations complicate efforts to support meaningful racial representation without reproducing stereotypes. We discuss how companion personalization should be evaluated beyond user satisfaction to account for broader representational harms.

Abstract: Open-vocabulary multimodal emotion recognition (OV-MER) aims to generate open natural-language emotion labels from multimodal affective cues. In real-world scenarios, however, complete and synchronized modal data are difficult to obtain due to limitations of acquisition devices and user privacy constraints. Existing OV-MER methods are largely designed for full-modal inputs, and fail to perform effective feature fusion under modal missing conditions. Meanwhile, current fusion approaches designed for incomplete modalities mainly focus on fixed-label recognition context, and cannot satisfy the demand for fuse emotional cues guided with arbitrary emotion semantics in OV-MER context. To tackle these challenges, this paper proposes an Affect-Prototype-Conditioned Fusion (APCF) framework for incomplete open-vocabulary emotion recognition. As a candidate-free generative framework, APCF extends modal contribution learning to scenarios guided by arbitrary emotional semantics. Specifically, we construct an affect-prototype library to explicitly model multimodal contribution characteristics corresponding to diverse emotions, which provides dynamic constraints for modal fusion under different emotional semantic perspectives. Conditional retrieval and feature aggregation are conducted based on available modal features. The refined fused affective representations are then fed into an LLM decoder to produce open-vocabulary emotion labels. Experiments on the OV-MERD+ and MER-FG datasets demonstrate that APCF substantially outperforms state-of-the-art baselines.

Abstract: Traditional recommender systems are typically trained to predict what item users will interact with next, but not why. However, offering personalized evidence for why a user might like the predicted item is an important way to enhance the service and to raise the likelihood that the user will be genuinely interested in the recommendation. This service can be delivered by integrating a frontier-model call into the member-facing pipeline, but it will add extra cost and latency. In this paper, we train a recommender LLM to generate personalized explanations for its reccomendation, based on the user's watching history at a large video streaming service. We impose two requirements on the generated explanation: it must be faithful to the elements of the shows it links, and it must be strictly non-harmful to the user. To this end, we first train two LLM-judge reward models covering three specific criteria, and propose constrained GRPO to incorporate these different criteria. On a held-out real-world testing set, our fine-tuned model improves the all-three-criteria PASS rate rises from 0.649 to 0.956 under our own judges and from 0.677 to 0.931 under an independent judge, where as the frontier generator performs similar to the untuned recommender baseline. We conduct further experiments to show that the model's language and recommendation abilities remain unchanged. Based on these results, we conclude that an LLM-based recommender can be fine-tuned on other complex tasks without compromising its original recommendation performance, thus provide insights for further agentic user interface powered by a single model.

Abstract: How often does a generative assistant persuade someone to buy, or persuade them not to buy? Conversation logs contain recommendations, but they do not necessarily record subsequent decisions. We audit 317 historical interactions from Aiso's proprietary research database of licensed, consent-based, de-identified conversations with commercially available AI assistants. Single-agent AI-assisted screening identifies 68 purchase-directed records; collapsing one shared-prefix copy yields 67 retained episodes, dated April 2023 to July 2025. Assistant responses provide candidate options, acquisition channels, or conditional preferences in 52 episodes (77.6%). One episode contains conditional redirection away from a named accommodation candidate. No episode is coded as advice to abandon or defer the purchase category. Only 18 episodes (26.9%) contain a subsequent user turn within the same purchase-related mission, compared with 23 (34.3%) that contain any later user turn. Using conversation depth alone therefore overstates this follow-up availability by 27.8%. Across 47 retained user follow-up messages, no explicit post-advice purchase commitment, completed-purchase report, or purchase-category abandonment statement is observed. These zeros describe recorded statements, not conversion or persuasion rates. The paper supplies operational definitions, text-free annotations, and reproducible descriptive results. Its central finding is a measurement limitation: recommendation content is observable much more often than a buyer's subsequent decision. The selected historical sample, unvalidated AI annotations, and missing transaction outcomes do not support a population-level or causal estimate of persuasion.

Abstract: The advance of multimodal large language models (MLLMs) has fundamentally reshaped the paradigm of human-computer interaction, especially speech interaction models capable of seamless conversations. Despite remarkable performance as general voice assistants, their performance in specialized domains remains underexplored, particularly in scientific areas. Scientific interactions introduce formidable challenges, involving rare technical terminology, spoken norms of abbreviations, and the natural verbalization of symbolic special expressions. In this paper, we introduce S$^3$-Bench, a systematic evaluation framework covering 10 major disciplines, consisting of a Knowledge set for speech question-answering and a Dialogue set for multi-turn progressive interactions with simulated user agents. By decomposing a complete atomic turn into stages of speech recognition, perception, knowledge utilization with reasoning, and response pronunciation, we systematically characterize the common challenges and performance tradeoffs of existing approaches. Furthermore, experiments on multi-turn interactions reveal persistent limitations in user adaptation and the generation of accurate, comprehensive, and efficient responses.

Abstract: Conversational Task Assistants (CTAs) are multimodal dialogue systems that support users in complex real-world tasks such as cooking and DIY through voice, text, image, and video interactions. Prior user studies have focused on controlled settings, leaving limited understanding of real-world CTA usage at scale. In this work, we present a large-scale study of CTA usage based on thousands of users in-the-wild. Our large-scale real-world data analysis unveils new understandings of (i) user-CTA interaction flows, (ii) user intents, (iii) user conversational traits, and (iv) behavioral factors associated with user satisfaction. Our findings reveal key opportunities for future research in CTAs, particularly in user interaction design and task engagement, concluding with concrete design guidelines.